A stages of change perspective onmanagers’ motivation to learn in aleadership development context

Stanley G. HarrisDepartment of Management, Auburn University,

Auburn, Alabama, USA, and

Michael S. ColeInstitute for Leadership and HR Management,University of St Gallen, St Gallen, Switzerland

Abstract

Purpose – The purpose of this paper is to examine the applicability of Prochaska and colleagues’“stages of change model,” which has generated substantial support in the therapeutic literature asa useful framework for understanding the dynamics of motivation to change problem behaviors, ina leadership development context.

Design/methodology/approach – A group of over 70 supervisors/managers was studied overa period of nine months as they participated in a company-sponsored leadership development effort.

Findings – Results provide initial evidence that the stages of change model has the potential forbeing reliably and validly assessed in a leadership development context. Participants’ stage scoresrelated in meaningful ways to relevant criteria such as job attitudes, perceptions of personal leadershipareas needing improvement, and evaluations of actual development module content and presentationover a nine-month period.

Research limitations/implications – Participants were drawn from only one organization andthis was the first major leadership development effort undertaken by this organization.

Practical implications – Study results provide support for the appropriateness of applying thestages of change model and its measurement in a leadership development context. Resultsdemonstrate that the stages of change model appears to offer useful and pragmatic insight intomotivation to learn and on improving the effectiveness of leadership development activities.

Originality/value – The present study is unique in that makes use of a stages of change model toempirically examine differential patterns of relationships between participants’ stages of change andtheir organizational attitudes, leadership developmental needs, and longitudinal reactions to thedevelopment effort.

Keywords Learning, Leadership development, Training, Managers, Leadership

Paper type Research paper

Increasingly, leadership development efforts are expected to play key roles inorganizations’ attempts to enhance their competitiveness and transform themselvesand their cultures (Armenakis et al., 1999; Tichy and Cohen, 1997). For example, themanagement and leadership development efforts of general electric are cited as keys toits competitiveness (Tichy and Sherman, 1993). More and more, leaders and managershave to be prepared to perform in team-based, global, fast-changing environments andorganizations have a vested interest in preparing them to do so (Goldstein and Ford,2002). Clearly, organizations have a keen interest in making sure that developmentefforts are as effective as possible.

The current issue and full text archive of this journal is available at

www.emeraldinsight.com/0953-4814.htm

JOCM20,6

774

Journal of Organizational ChangeManagementVol. 20 No. 6, 2007pp. 774-793q Emerald Group Publishing Limited0953-4814DOI 10.1108/09534810710831019

Motivation to learn is recognized as playing an important role in the ultimatesuccess of training and development activities (Baldwin et al., 1991; Tannenbaum andYukl, 1992). Generally, motivation to learn is conceptualized as exerting its influencethrough a participant’s decision-making process regarding the direction, focus, andlevel of their effort to participate in the developmental activity (Noe et al., 1997). Inshort, people are more likely to learn and apply that learning when they are motivatedto do so. After reviewing the literature, Colquitt et al. (2000, p. 681) concluded “there is arobust positive relationship between motivation to learn and learning outcomes”.

Kirkpatrick’s (1994) classification of learning outcomes into reaction, learning,behavior, and results has proven very useful for both training practitioners andresearchers (Goldstein and Ford, 2002). Reaction refers to evaluations the participantsmake of the training/development program itself. Satisfaction with the instructor,materials, course structure and process are examples of reactions (Goldstein and Ford,2002). Kraiger et al. (1993) have divided the learning outcome into three subcategories:cognitive (e.g. acquiring new knowledge), skill based (e.g. mastering a new behavior),and affective (e.g. improved self-efficacy). The behavior outcome refers to training- ordevelopment-specific job performance. Finally, results refer to the influence of thetraining on organizational-level outcomes such as costs, turnover, and productivity.

In theoretical conceptualizations of the training process (Colquitt et al., 2000;Goldstein and Ford, 2002), trainee learning motivation and ability (particularlycognitive), along with situational variables, are treated as predictors of trainee learningand reactions to the training. These learning outcomes, in turn, are assumed to predictlearning transfer and performance. Research generally supports this theoreticalframework. Most notably, in their meta-analysis of 106 research studies, Colquitt et al.(2000) reported that motivation to learn explained significant variance in all measuredlearning outcomes even after accounting for cognitive ability. They reportedsignificant meta-analytic path coefficients from learning motivation to cognitivelearning (0.39), skill acquisition (0.22), post-training self-efficacy (0.22), and reactions tothe training (0.45). In turn, skill acquisition and post-training self-efficacy hadsignificant paths with knowledge transfer (0.59 and 0.27 respectfully) and knowledgetransfer had a significant path coefficient of 0.59 with job performance.

Despite the statistical strength of the relationships between learning motivation andtraining outcomes, we suggest that understanding motivation to learn might beenriched by approaching it from yet another theoretical framework offered by thechange management literature. Since, the decision to actively participate in adevelopmental activity implies a desire or willingness for personal change, a focus onthe dynamics of willingness to change provides an alternative perspective onmotivation to learn. Such a perspective may help with three issues facing themotivation to learn literature. First, while significant, the magnitude of the relationshipbetween learning motivation and training outcomes is oftentimes modest. For example,Baldwin et al. (1991), Colquitt and Simmering (1998) and Mathieu et al. (1992) reportedthat pre-training motivation explained only 2 percent of the incremental variance inparticipants’ post-training learning. In the results from Colquitt et al. (2000) reportedabove, learning motivation accounted for between 5 and 15 percent of the variance inlearning outcomes. Perhaps, a theoretical expansion of learning motivation beyondtraditional motivational conceptualizations to include willingness for personal changecould improve its predictive powers.

Stages of changeperspective

775

Second, in their meta-analysis, Colquitt et al. (2000) found that the influences oforganizational climate and the individual characteristics of locus of control,conscientiousness, anxiety, and age on training outcomes were only partiallymediated through learning motivation. This led them to ask, “If the effects ofindividual and situational characteristics are not fully mediated by self-efficacy,valence, job/career variables, and motivation to learn, what are the other interveningmechanisms?” (p. 701). Change willingness might be one such mechanism.

Finally, Cronbach and Snow (1977) discussed the importance of anaptitude-treatment interaction which emphasizes the matching of instructionalmodel with trainee aptitude. Baldwin and Ford (1988) posited that a match was alsoneeded with trainee personality (personality-treatment interaction). Since, learningmotivation is an important factor in the success of training and development programs,it follows that efforts must be made to tailor training to the motivations of theindividuals and engage in pretraining interventions designed to improve motivation(Goldstein and Ford, 2002). Much of the literature on personal and organizationalchange emphasizes just such interventions and should prove useful in guiding trainingand instructional design decisions. Therefore, a change willingness/readinessapproach might suggest the importance of a developmental readiness-treatmentinteraction.

With organizations continuing to spend significant amounts of resources on theirtraining and development programs (Thompson et al., 2002), it seems prudent toconsider theoretical approaches from disciplines outside traditional training anddevelopment theory that might yield new insights into motivation to learn. Cole et al.(2004) have demonstrated that a theory focusing on the motivational dynamics ofdiverse stages of personal change readiness, developed and championed by Prochaska(1979), offers a unique and potentially useful framework from which to explore themotivation of participants to engage in developmental change by active involvement ina development program. Prochaska’s work was originally developed in the context ofpsychotherapy and therapeutic efforts to encourage positive change such as smokingcessation and condom use. Recent work has extended Prochaska’s model toorganizational phenomena, including readiness for organizational change(Cunningham et al., 2002; Prochaska et al., 2001b).

Cole et al. (2004) utilized the Prochaska’s model to develop and test a stage-basedapproach to measuring motivation that could be applied to specific university courses.They found that the stages of change approach predicted training outcomes even aftercontrolling for a traditional measure of motivation to learn. Building off theseextensions, we argue that there are insightful parallels to be drawn between readinessfor change in therapeutic settings and college classrooms and readiness for personalleadership development.

To examine the applicability of Prochaska’s stages of change model (Prochaskaet al., 1992, for a review) for understanding managers’ motivation to learn in anorganizational developmental context, we studied a group of over 70 supervisors andmanagers over a period of nine months as they participated in a company-sponsoredleadership development effort. Based on a review of the stages of change model, wearticulate and test five hypotheses concerning the relationship between the stages andparticipants’ learning motivation, readiness for development, organizational attitudes,

JOCM20,6

776

perceptions of their need for development, and attitudinal reactions to the programcontent and delivery.

Prochaska’s stages of change modelIn an attempt to understand the successes and failures of both self-mediated andpsychotherapeutically-facilitated efforts to intentionally change problem behaviorssuch as smoking and overeating, Prochaska (1979) emphasized the role of motivationto change and articulated a “transtheoretical model of change” incorporatingmotivational, cognitive, social learning, and relapse prevention theories. In an earlystudy guided by this change model, DiClemente and Prochaska (1982) compared theexperiences of individuals who had stopped smoking on their own with individualsparticipating in treatment programs. Interestingly, they found strong evidence acrossboth samples for progression through distinct stages of change marked by differentmotivations, concerns, and intervention requirements. These findings led to a series ofstudies whose results were articulated in a stages of change model (DiClemente andProchaska, 1982; Prochaska and DiClemente, 1982, 1983) that provides a framework forcategorizing stages of change readiness and motivations and ultimately theinterventions appropriate to support progress at each stage.

In the first stage of the model, precontemplation, individuals are either naivelyunaware of the need for change or actively resist awareness of the need (Prochaska andDiClemente, 1982) and have no intentions of changing in the foreseeable future(Prochaska et al., 1992). In the second stage, contemplation, individuals are aware of aneed for change and are seriously considering changing but have not yet committed toit (Prochaska et al., 1992). At the third stage, preparation, individuals are intending totake action very soon or have already initiated initial actions. At the fourth stage,called “action,” contemplation and preparation gives way to actual initiation of changebehaviors. In the fifth and final maintenance stage, motivation is focused on theprevention of relapse to the pre-change set of behaviors (McConnaughy et al., 1989;Prochaska et al., 1992). Using the decisional balance inventory (Velicer et al., 1985),Prochaska et al. (1994) demonstrated that the stages were related to the relativeweighting of the perceived pros and cons of the change. In precontemplation, the consdominated but started to deteriorate at higher stages with the pros dominating at theaction stage.

The stages of change model has received an extraordinary amount of empiricalevidence supporting its ability to predict behavior change across a variety of problembehaviors. In an integrative study investigating the generalizability of the stages ofchange model, Prochaska et al. (1994) examined a wide range of personal changebehaviors across a variety of samples. In all, they examined 12 behaviors: smokingcessation, quitting cocaine, weight control, reducing dietary fat, reducing adolescentdelinquent behaviors, practicing safe sex, condom use, sunscreen use, avoiding radongas exposure, exercise acquisition, mammography testing, and physicians’ preventivepractices with smokers. The authors purposefully chose a wide variety of problemsthat differed in many ways, including their frequency of occurrence, legality of thebehavior, the visibility of the behavior (i.e. public vs private setting), and socialacceptance of the action. In addition, some changes involved ceasing problembehaviors (e.g. smoking and cocaine use) while others involve practicing safe behaviors

Stages of changeperspective

777

(e.g. condom use and mammography testing). The researchers found the stages ofchange model was confirmed across each sample.

Since, its introduction, the stages of change model has generated research resultswith pragmatic implications for understanding and reacting appropriately to thedifferent types of readiness and willingness for change associated with each stage. Inessence, this research suggests that “efficient self-change depends on doing the rightthings (processes) at the right time (stages)” (Prochaska et al., 1992, p. 1110). Forexample, in the Prochaska et al. (1994) study of 12 different behaviors, the researchers,based on the respondents’ change stage, were able to accurately and reliably predictthe decisional processes and types of techniques that facilitated individuals to makeprogress.

Not surprisingly, the stages of change model has started finding its way intoorganizational research. Prochaska et al. (2001b) provide a good overview of much ofthis work. In their study of a large-scale change effort in a hospital, Cunningham et al.(2002) provide preliminary support that both individual (e.g. self-efficacy) and jobcharacteristics (e.g. decision latitude) variables were positively related to changereadiness, as operationalized by the stages of change, and readiness was positivelyrelated to subsequent participation in the change activities.

Prochaska et al. (2001a, b) have summarized the fundamental processes that can beused to motivate organizational change and their appropriate application to matchreadiness stages. Specifically, they argue that consciousness raising (increasingawareness of the problem or need for change), dramatic relief (creating fear of notchanging or excitement for changing), and environmental reevaluation (creating anappreciation of the social and work environment advantages of the change) are mostappropriate when individuals are in precontemplation. To move individuals from theprecontemplation to contemplation stages, they suggest that efforts should focus onreducing perceptions of the cons for change and increasing perceptions of the pros.Self-reevaluation in order to create an appreciation of the positive benefits of thechange for self identity, happiness, and success is deemed most appropriate forthe contemplation stage. To move from contemplation to preparation, they suggest theimportance of taking small behavioral steps. Self-liberation strategies focused onincreasing efficacy regarding successfully changing best match the preparation stage.Finally, more behavioral, skill, reinforcement, and action-oriented change processesseem best suited for the action and maintenance phases. Specifically, they judgereinforcement, helping relationships through social support, counter-conditioning(replacing old ways with new behaviors and cognitions), and stimulus control(changing the environment to facilitate new behaviors and inhibit the old) as mostappropriate for the action and maintenance stages.

As demonstrated by Cole et al. (2004) in college classrooms, the stages of changemodel appears to also be highly applicable to management learning contexts. Inparticular, they provide results suggesting that action sentiments are most consistentwith traditional treatments of learning motivation. In other words, learning motivationassumes an awareness of a need for change. The stages of change approach augmentsthat assumption by emphasizing the importance of awareness that accompaniesmovement from precontemplation to contemplation. It seems likely that potentialparticipants in a management development activity would have differential degrees ofawareness of the need for, and desires to participate in, such development.

JOCM20,6

778

Furthermore, if participants’ stages of readiness were assessed reliably, more precisetailoring of developmental content and approaches to match the stage needs ofparticipants would be facilitated. The study described here is a first attempt to applythe stages of change model to a company-sponsored leadership development context.

HypothesesIf the stages of change model offers a valid approach to understanding pre-trainingmotivational sentiments, then it should demonstrate appropriate convergent anddiscriminant validity and relate in predictable ways with other organizational andtraining relevant variables. Specifically, in the present study, we propose that thestages of change should relate in meaningful ways to participants’ developmentalreadiness, learning motivation, organizational attitudes, perceived developmentalneeds, and reactions to the developmental program content and delivery.

In the remainder of this paper, we limit our focus to three of the five change stages:precontemplation, contemplation and action. Preparation is oftentimes treated andmeasured more as a transition phase between precontemplation and contemplationrather than a distinct stage. Prochaska et al. (1992) have suggested that preparationsentiments are inferred from scores on contemplation and action scales, and, therefore,do not explicitly measure preparation. We also felt the omission of the maintenancestage was appropriate since the focus of leadership development is generally onacquiring new skills and abilities rather than providing support and encouragementfor those already acquired and practiced. This omission is consistent with the empiricaldecisions made by others (Cole et al., 2004; Lam et al., 1988).

Readiness for developmental changeProchaska et al. (2001b), clearly equate progression through the stages of change withincreased readiness for change. For example, Prochaska et al. (2005) use the stagemodel to assess emotional readiness to adopt a child. Cunningham et al. (2002) go so faras to measure readiness with items directly assessing each of the five stages of change.Outside the stages of change context, the concept of readiness for change has enjoyed agreat deal of attention in the organizational sciences. Armenakis et al. (1999, 1993)outlined the role of readiness in precluding resistance and encouraging changeacceptance and institutionalization. In addition, they have outlined ways thatorganizations can increase readiness. Armenakis et al. (1999) argue that readiness forchange is the result of five sentiments: change is needed, the particular change beingconsidered is appropriate, the change is doable (efficacy), the organization is committedto the change, and the change has personally-relevant benefits. These sentiments are ofa different sort and more extensive than those addressed in the stages of change model,which typically focus broadly on the balance of pros and cons. However, we wouldexpect that the stages of change would relate to readiness as assessed from theArmenakis et al. (1999, 1993) perspective. Specifically:

H1. Precontemplation will negatively correlate with readiness for developmentalchange and contemplation and action will positively correlate.

Learning orientationLearning orientation is an individual disposition that is associated with a predispositionto want to learn and increase competence (Colquitt and Simmering, 1998). In the context

Stages of changeperspective

779

of a development effort, one would expect that learning orientation would encourageconstant contemplation of opportunities for development and involvement in suchactivities. In fact, Colquitt and Simmering (1998) found that learning orientationpredicted motivation to learn in a classroom setting, even after controlling for trainingexpectancy and valence. Therefore, we expect the following:

H2. Learning orientation will correlate negatively with precontemplation andpositively with contemplation and action.

Organizational attitudesPerceived organizational support refers to employees’ feelings that the organizationvalues them and cares about their well-being (Eisenberger et al., 1986). Affectivecommitment refers to the emotional identification with the organization and its valuesand culture (Meyer and Allen, 1984). Social exchange theory (Blau, 1964) and the normof reciprocity (Gouldner, 1960) suggest that such positive feelings about theorganization encourage reciprocity on the part of the employee. One form of reciprocitymight include support for development activities provided by the organization.Therefore, we expect that participants with more positive attitudes toward theirorganization (i.e. affective commitment and perceived organizational support) will bemore predisposed to consider (contemplate) and ultimately embrace developmentalopportunities afforded them by the organization. On the other hand, individuals whohold negative organizational attitudes are more likely to be skeptical of the motivesbehind any developmental offering. Therefore, we offer the following hypothesis:

H3. Perceived organizational support and affective organizational commitmentwill correlate negatively with precontemplation and positively withcontemplation of the developmental opportunity.

Developmental needsBy definition, the extent to which a person feels they have developmental needs directlycorresponds to their precontemplation and contemplation sentiments regarding adevelopmental opportunity. In effect, awareness of need requires contemplation.Therefore, any measurement of the stages of change should demonstrate convergentand discriminant validity such that:

H4. Precontemplation will correspond to lower levels of felt developmental needswhile contemplation will correspond to higher levels of felt needs.

Reactions to the development programIf the stages of change approach is to contribute to our understanding of thedevelopment process, it must relate to important development outcomes. InKirkpatrick’s (1994) classification of learning outcomes, participant reaction towardthe development program is an important first-level contributor to other outcomes. Weexpect that participants’ stages of change sentiments relative to the leadershipdevelopment program will explain variance in their reactions to the program,particularly satisfaction with the instructional content and delivery of specificdevelopment modules. As noted by Alliger et al. (1997), positive reactions to a trainingsituation are important because satisfaction might have substantial influence on othertraining outcomes such as attendance, performance, or “word-of-mouth” advertising

JOCM20,6

780

that might impact future participants’ involvement. In their study of college classes,Cole et al. (2004) found that students in the precontemplation stage were absent moreoften and less satisfied at the end of the class than others. Individuals who arerecipients of developmental activities that they feel are unneeded or unwanted mighteasily become resentful and critical of the program content and delivery. Therefore, wewould expect precontemplation to relate negatively with the participants’ reactions tothe program. Likewise, if a person is contemplating change or actively involved in it,they are more likely to have positive reactions to the development effort:

H5. Precontemplation will correlate negatively with evaluations of thedevelopment program content and delivery, while contemplation and actionwill correlate positively.

MethodResearch setting and sampleWe conducted our study in a large manufacturing division of a multinationalcorporation. All division operations occupied one large campus in the southeasternUSA. At the request of the division’s top management team, a long-term leadershipdevelopment program was created and implemented by the business school from anearby university. The program was designed to be delivered in approximately nineone-day modules (each module offered three separate times to facilitate participantscheduling) over a nine-month period. Modules covered topics such as self-awareness,corporate strategy, finance, change management, communication, and quality control.This was the first such developmental effort engaged in by the division.

While not formally mandated, division executives strongly encouragedparticipation by all supervisors (N ¼ 35), managers (N ¼ 39), and select engineeringstaff with project management duties (N ¼ 8). Therefore, the sample for the presentstudy consisted of approximately 82 eligible program participants encouraged toattend one of the three session offerings of the first development module. It is possiblethat participants felt little choice regarding attending the development sessions. Since,realized choice is associated with greater motivation to learn (Baldwin et al., 1991), itmight be argued that a no-choice option serves the present research best sinceparticipants are likely to have a greater range of motivation for the developmentprogram. Of those attending the first module, 74 volunteered to participate in the datacollection efforts. Because of the relatively small sample and to further encourageparticipation and ensure anonymity, no demographic data were collected.

Data collectionThe research reported here was conducted over a span of nine months and covered thenine distinct developmental modules. During this period, participants were giventhe opportunity to provide data on several occasions. All data were anonymous. To beable to match participants’ responses over time, participants were asked to developtheir own personal code that they could supply with each submission of data.Participants could also choose to turn in information without supplying a code.

On the initial developmental module session date, and prior to any presentation ofdevelopmental content, all attendees were asked to complete a questionnaire thatincluded our measure of the stages of change and items regarding their perceivedneed for training in certain areas, their dispositions, and organizational attitudes.

Stages of changeperspective

781

A cover letter explaining the purpose of the survey, assurances of anonymity, and therequest that they develop and use a personal code was also provided.

We also collected data in the form of evaluations at the end of each of the ninetraining sessions conducted during the period covered by the study. Participantswere asked to provide an evaluation of both the content and delivery (instructor) of themodule material. Participants were encouraged to write their code on the evaluationforms.

MeasuresUnless otherwise noted, all measures employed statements to which respondents wereasked to indicate the degree to which they agreed with them on a five-point Likertscale (1 ¼ strongly disagree; 5 ¼ strongly agree). All measures, with the exception ofthe developmental module evaluations were collected immediately prior to thepresentation of the initial developmental module.

Stages of change scale (SOCS). To develop a measure that discretely captured thestages of change for the leadership development context, we were guided by validatedmeasures from the therapeutic literature. We wanted a scale that would provideseparate measures for each change stage. Accordingly, we chose to modify 12 itemspreviously validated by Lam et al. (1991) and McConnaughy et al. (1989). Followingprevious convention (Lam et al., 1988), our stages of change scale (SOCS) did notinclude items for either the preparation or maintenance stages.

A principal-components analysis with varimax rotation was used to examine theunderlying dimensionality of the SOCS. Based on the scree plot,eigenvalues-greater-than one, and factor loadings (.0.45), results of theprinciple-components analysis suggested three underlying factors were present in thescale, explaining 64.6 percent of the variance. Shown in Table I, the three factors closelymatched those of Prochaska and colleagues’ conceptualization of the stages of changemodel. Based on these results, the items representing each stage were averaged together(items 11, 19, and 20 were reverse scored) to create scale scores. Factor 1, action,consisted of five items and had a coefficient a reliability of 0.86. Factor 2, contemplation,was made up of three items and had a coefficient a reliability of 0.75. Factor 3,precontemplation, consisted of four items and had a coefficient a reliability of 0.72.

To be consistent with Prochaska’s (1979) theory, the SOCS stage scores shouldexhibit a simplex structure, whereby adjacent stage scores correlate more stronglythan non-adjacent scale scores. In addition, precontemplation sentiments would beexpected to correlate negatively with the other more change-supportive stages. Thecorrelations among the three scales supported the simplex structure assumption.Specifically, precontemplation was negatively correlated with contemplation(r ¼ 20.55, p , 0.01) but was not correlated with action (r ¼ 20.08, ns). Asexpected, contemplation was positively correlated with action (r ¼ 0.31, p , 0.01).This pattern of relationships is consistent with that exhibited in the original studiesmeasuring the stages reported by McConnaughy et al. (1983, 1989) and supports thevalidity of our SOCS.

Change readiness. We assessed change readiness using a six-item scale based on thework of Armenakis et al. (1993). Paralleling the theoretical foundations of motivation tolearn, their change readiness concept is derived from expectancy theory, and suggeststhat readiness depends on a perceived need for change (two items: e.g. “My leadership

JOCM20,6

782

skills need improving”), efficacy with regard to making the change (two items: e.g.“If I try, I can become a better leader”), and assessment of the personal valence ofchange (two items: e.g. “Becoming a better leader is important to me”). Internalconsistency for this scale was 0.87.

Learning orientation. Six items from VandeWalle’s (1996) learning orientation scale(e.g. “I often look for opportunities to develop new skills and knowledge”) included inhis goal orientation inventory were used to assess learning orientation. Internalconsistency for the scale was 0.81.

Perceived organizational support. Participants were asked to indicate theiragreement with 16 items from Eisenberger’s perceived organizational support scale(Eisenberger et al., 1986) (e.g. “The company regards my best interests when it makesdecisions that affect me” and “Help is available from my organization when I have aproblem”). The internal consistency for this scale was 0.94.

Affective organizational commitment. Commitment was measured with the affectivecommitment scale, used in research by Meyer and Allen (1984; Allen and Meyer, 1990).

Factor

Scales and items Eigen-valueVarianceexplained I II III

Precontemplation 3.35 27.91. As far as I am concerned, I do not have anyleadership development needs 20.73911. I am hoping this leadership developmentprogram will help me to better understand myselfa 0.63419. Maybe this leadership development program willbe able to help me become a better leadera 0.59820. I hope that I get some good advice from thisleadership development programa 0.584Contemplation 2.51 21.014. I have some leadership challenges and I reallythink I should work on them 0.7295. It might be worthwhile to work on improving myleadership skills 0.7262. I think I might be ready for some leadershipself-improvement 0.710Action 1.9 15.76. I have been working on improving my leadershipskills 0.88021. I am actively working on my leadershipshortcomings 0.8373. I am doing something about my leadershipshortcomings that have been bothering me 0.7909. I am really working hard to improve myleadership skills and approach 0.77912. Even though I am not always successful inchanging, I am at least working on improving myleadership skills 0.679

Notes: aItems were reverse scored. Scales were computed as the average of their items

Table I.Results from the

principal-componentsanalysis

Stages of changeperspective

783

Researchers using the affective commitment scale have reported that its items (e.g. “Ifeel a strong sense of belonging to the company”) form a single factor with highreliability (Allen and Meyer, 1990). Internal consistency for this scale was 0.83.

Developmental needs. Participants were asked to indicate how much personalimprovement they needed to make with regard to 24 specific leadership areas that weregrouped into six distinct leadership skill domains (four items per domain). Participantsresponded using a five-point scale (1 ¼ hardly any improvement needed; 5 ¼ very greatimprovement needed). Scale scores for each leadership skill domain were created byaveraging participants’ responses. The six leadership skill domain scales were: managingconflict (a ¼ 0.91), budgeting (a ¼ 0.92), quality management (a ¼ 0.86), visionaryleadership (a ¼ 0.87), teambuilding (a ¼ 0.72), and stress management (a ¼ 0.73).

Developmental module evaluations. Scales were developed to evaluate the contentand instruction of each of the training modules conducted during the nine month study.A team of two instructors conducted one training module (change management) but aseparate evaluation covering both content and instructor was conducted for eachinstructor of this module. One instructor (instructor 1) conducted three of the modules.All other instructors participated in only one module each. The content evaluation ofeach module was assessed by a scale created using three items (e.g. “I can use thisinformation” and “I would recommend these seminars”). The instructor evaluation ofeach module was assessed using six items addressing the instructors’ knowledge,communication skills, preparedness and organization, and motivational skill (e.g. “Theinstructor communicated well with the class” and “The instructor created interest andstimulated thinking”).

Controls. Because our variables are attitudinal and collected using a commonmethod, inflation in the relationships among them due to common method anddispositional-based response tendencies becomes a particular concern (Spector andBrannick, 1995). Therefore, we measured three dispositional variables to use ascontrols in the analyses: positive (a ¼ 0.90) and negative (a ¼ 0.90) emotionality(Tellegen, 1982, 1985) and self-deception (a ¼ 0.69, Paulhus, 1991).

Data analysis strategyOur first four hypotheses involved investigating the relationships betweenparticipants’ SOCS scores and other attitudinal variables. To combat commonmethod concerns, partial correlations were examined, controlling for positive andnegative emotionality and self-deception. Finally, to examine the relationships betweenthe SOCS and the attitudinal outcomes of training that were the focus of H5, simplecorrelation coefficients between participants’ stage scale scores and their reactionswere investigated.

ResultsThe relationships between our measures of the precontemplation, contemplation, andaction stages and the criterion variables for H1-H4 were assessed using third-orderpartial correlations. The results of these analyses are shown in Table II.

Hypothesis 1As shown in Table II and consistent with H1, all three stage scores were correlated tothe change readiness measure. Precontemplation sentiments were negatively

JOCM20,6

784

correlated with being ready to change (r12.345 ¼ 20.60, p , 0.01). In contrast,readiness was positively correlated with contemplation (r12.345 ¼ 0.70, p , 0.01), andaction (r12.345 ¼ 0.28, p , 0.01) scores.

Hypothesis 2Contrary to H2, learning orientation was not associated with either precontemplationor contemplation; however, it was related to action sentiments (r12.345 ¼ 0.33, p , 0.01).Thus, people who have a learning orientation may be more likely to have alreadymoved beyond contemplation and into action, particularly with regard to their ownleadership development, even before a formal development activity is offered.

Hypothesis 3Consistent with H3, perceived organizational support was marginally negativelyrelated to precontemplation (r12.345 ¼ 20.17, p , 0.10) and positively withcontemplation (r12.345 ¼ 0.18, p , 0.10) scores. Stronger relationships were foundbetween affective commitment and precontemplation (r12.345 ¼ 20.23, p , 0.05)and contemplation (r12.345 ¼ 0.21, p , 0.05). Neither organizational attitude wasassociated with action scores. It appears that positive organizational attitudes may,indeed, encourage more contemplation of developmental opportunities offered by theorganization.

Hypothesis 4Consistent with H4, precontemplation scores exhibited negative relationships( p , 0.05) with perceived developmental need for five of the six developmental areas.Only stress management was not related to precontemplation. Similarly, contemplationscores exhibited positive correlations ( p , 0.01) with the same five development needareas. As hypothesized, regardless of statistical significance, the precontemplation scalehad negative partial correlations with the perception of all need areas (averager12.345 ¼ 20.31) while contemplation had positive correlations (average r12.345 ¼ 0.35)with all. None of the six developmental needs were related to action scores.

Stages of change scalesPrecontemplation Contemplation Action

Change readiness 20.60 * * 0.70 * * 0.28 * *

Learning goal orientation 0.09 0.10 0.33 * *

Organizational attitudesPerceived organizational support 20.17 0.18 0.14Affective organizational commitment 20.23 * 0.21 * 0.06

Perceived developmental needsManaging conflict 20.37 * * 0.44 * * 0.11Budgeting 20.40 * * 0.39 * * 20.02Quality management 20.47 * * 0.38 * * 20.06Visionary leadership 20.20 * 0.34 * * 0.05Team building 20.30 * * 0.35 * * 2 0.03Stress management 20.09 0.19 0.14

Notes: *p , 0.05; * *p , 0.01. Ns range from 68 to 69. aAfter controlling for positive and negativeemotionality and self-deception

Table II.Partial correlationsa

between stages of changeand learning orientations,

organizational attitudes,and perceived

developmental needs

Stages of changeperspective

785

This is entirely consistent with expectations created by stages of changeperspective on motivation for development: precontemplative sentimentsare associated with less perceived need for change while contemplation sentimentsare associated with higher needs for change. Furthermore, action sentiments suggestindividuals are currently hard at work to improve, and, therefore, they may feel furtherchanges are not needed.

Hypothesis 5Development modules were constructed and, for nine consecutive months, adevelopment module was offered to eligible company employees. At the end of eachdevelopment session, participants were asked to evaluate the module’s content and theinstructor providing the training. In regard to participants’ evaluations, H5 predictedthat participants’ initial scores on the SOCS will be associated with subsequentjudgments concerning development modules’ adequacy and instructors’ competence.

Correlations between the SOCS and participants’ content and instructor evaluationsfor the nine developmental modules are reported in Table III. Zero-order correlationswere examined out of a consideration for statistical power. Had we chosen partialcorrelation analysis, additional cases would have been excluded from the analysisresulting in an even smaller sample size and further loss of statistical power. UsingFisher’s zr, we compared the magnitude of the zero-order correlations reported inTable III with correlations obtained after controlling for participants’ emotionality andself-deception. In every instance, Fisher’s zr indicated there were no statistically

Stages of change scalesVariables N Precontemplation Contemplation Action

Content evaluationsSession 1 (self-awareness) 59 20.09 20.14 20.11Session 2 (strategic analysis) 37 20.16 0.19 20.07Session 3 (strategic controls) 36 20.12 0.16 20.04Session 4 (quality management) 39 20.45 * * * 0.44 * * * 0.04Session 5a (change management) 34 20.24 * 0.26 * 0.23 *

Session 5b (change management) 31 20.34 * * 0.28 * 0.23Session 6 (communication) 37 20.16 0.17 0.24 *

Session 7 (financial measures) 21 20.34 * 0.28 0.18Session 8 (team building) 35 20.39 * * 0.45 * * * 0.48 * * *

Session 9 (stress management) 30 20.35 * * 0.33 * * 0.21Instructor evaluationsSession 1, instructor 1 60 20.01 20.06 20.02Session 2, instructor 2 37 20.23 * 0.24 * 20.09Session 3, instructor 3 36 0.06 0.07 0.07Session 4, instructor 4 39 20.30 * * 0.40 * * * 20.07Session 5a, instructor 1 34 20.08 0.30 * * 0.27 *

Session 5b, instructor 5 31 20.06 0.01 0.11Session 6, instructor 6 37 0.07 20.07 0.14Session 7, instructor 7 21 0.02 0.21 0.07Session 8, instructor 1 35 20.11 0.18 20.01Session 9, instructor 8 30 20.13 0.22 20.11

Notes: *p , 0.10; * *p , 0.05; * * *p , 0.01

Table III.Correlations betweenstages of changeand development moduleevaluations

JOCM20,6

786

significant differences between the correlations’ magnitudes. Because our initialsample size was further diminished because several original participants failed toprovide identification codes on their module evaluations, we tested for statisticalsignificance at the 0.01, 0.05, and 0.10 levels.

The pattern of results shown in Table III is clear and consistent with H5:precontemplation is associated with negative perceptions of module content andinstruction while contemplation is associated with positive perceptions.Precontemplation scores exhibited negative correlations ( p , 0.10) with the contentratings of six of the ten content evaluations and two of the ten instructor ratings.Contemplation scores were positively correlated ( p , 0.10) with content evaluations offive modules and three of the instructor ratings. Only the content evaluation of theteam building module (session 8) was strongly ( p , 0.01) correlated with action scores.Two additional content evaluations and one instructor rating were moderately( p , 0.10) related to action scores.

DiscussionStudy contributionsThis study provides some initial evidence that the use of “. . . one of the most influentialmodels in the area of health behavior change within the last 20 years” (Morera et al.,1998, p. 39) has the potential for being reliably and validly assessed in a leadershipdevelopment context and offers insight into other attitudes bearing on motivation tolearn and reactions to development experiences. Upon modification of the stages ofchange items to correspond with a leadership development context, results indicatedthe items loaded onto one of three factors consistent with Prochaska’s initialconceptualization. In addition the SOCS scales had high-internal consistency andparticipants’ precontemplation and contemplation scale-scores exhibited acceptablediscriminant and convergent validity with regard to the most of criterion variables. Insum, the SOCS shows promise as a measure that can be reliably and validly used indevelopment contexts.

In general, the pattern of relationships among the variables was consistent with ourhypotheses. More specifically, participants’ precontemplation scores were negativelyassociated with criterion measures of change readiness, affective organizationalcommitment, leadership improvement areas, and ratings of the program. It isimportant to note at this point that even those relationships that were non-significantwere generally in the expected direction (this point is particularly noteworthy giventhat, with the low-sample size and corresponding low-statistical power, correlations ashigh as 0.34 were only marginally significant). These results imply thatprecontemplation sentiments are associated with lower organizational commitment,lower perceived developmental needs, and more negative evaluations of actualdevelopmental content and instructors. Conversely, participants’ contemplationsentiments were associated with greater organizational commitment, perceiveddevelopmental needs, and more positive evaluations of actual developmental contentand instructors.

These results provide credence to warnings from the therapeutic world thatprecontemplation and contemplation refer to different motivations for change and thatchange efforts must be tailored to each (Prochaska et al., 1992). Based on participants’reactions to the diverse developmental modules offered during the span of this study,

Stages of changeperspective

787

it is clear that greater precontemplation sentiments led to harsher evaluations whilegreater contemplation sentiments led to more favorable ones. This would suggest thatthe leadership development program was geared primarily to meet the needs ofcontemplators. The majority of the material presented was conceptual in nature andrelated to information participants could use to work on improving their leadershipbehaviors. The content generally assumed that participants would value it and littleeffort was made to “sell” the content as important and needed. Luckily, contemplationsentiments dominated in the program participants. But what if they had not?

Our research suggests an extension to Cronbach and Snow’s (1977) aptitude-treatmentinteraction and Baldwin and Ford’s (1988) personality-treatment interaction.Aptitude-treatment argues that instruction should be tailored to individual aptitude.Personality-treatment makes the same argument for tailoring for personality. Similarly,our work suggests the importance of developmental readiness-treatment interaction.

The stages of change model would suggest that interventions designed forcontemplators will not work on precontemplators. Therefore, the first step is to get allmembers of the audience to move to the contemplation stage. What this means is thatproviders of leadership development opportunities should make sure thoseopportunities occur after participants are ready for them. How can precontemplatorsbe moved to contemplators? The key is raising awareness of a need for change anddevelopment (Prochaska et al., 1992). This implies that the first step of any leadershipdevelopment effort should involve consciousness raising and diagnosis of the need forchange through special workshops or other preparatory initiatives. While currentcontemplators may feel such efforts are “not needed” and become frustrated withwaiting for the development they desire, at least the whole development program willnot be a wasted effort in terms of those with precontemplation sentiments. Such is thekind of insight that we feel the stages of change model has to offer the managementlearning community.

One can also conclude that this study suggests the importance to leadershipdevelopment efforts of an organization that promotes self-evaluation, performancestandards, and an aligned performance management system. We found thatsentiments such as perceived organizational support and affective commitmentlikely to stem from such organizational activities would make precontemplationsentiments difficult and would instead promote constant contemplation and action.The importance of considering organizational support for change is exemplified inLevesque et al.’s (2001) research assessing the stages of change at the organizationallevel with regard to involvement in continuous quality improvement among VeteransHeath Administration hospitals. One interesting finding from their work was thathospitals characterized as being in the action or maintenance stages sent moreparticipants to a quality-related training program those in prior stages.

Contrary to our expectations, we did not find a relationship between participants’learning orientation and precontemplation or contemplation. Instead, learningorientation was related only to the action stage. Seemingly, those with a learningorientation are likely to already taking action with regard to their own development. Itis interesting to note that this pattern of relationship mirrors the relationships betweenthe stages and motivation to learn reported by Cole et al. (2004) and adds additionalsupport to the idea that the stages of change add a unique perspective on readiness andwillingness to embrace a developmental offering.

JOCM20,6

788

Study limitations and future research opportunitiesAdmittedly, the present study has limitations. One of the most basic limitations is thesmall sample size. This problem was exacerbated when we used participants’longitudinal evaluations as the criteria, since sample sizes were reduced furtherbecause many participants failed to provide an identification code on their evaluationforms. As a result of the low-statistical power of our analyses, the significant resultsobtained provide conservative evidence of the underlying relationships.

Another significant shortcoming of the present research is that participants weredrawn from only one organization and one developmental effort. This was the firstmajor leadership development effort undertaken by this organization. The resultsmight be different in an organization with more developmental history. Inorganizations engaged in such efforts over time, one would expect that reactionswill also be biased by assessments and perceptions of prior activities. The developmenteffort we studied also cast a rather large net with regard to participants. Participantswere drawn from the population of managers and supervisors without regard forperformance. Not surprisingly, we found a range of readiness stage scores. This rangewould likely be restricted for development efforts targeted only to “high potential”candidates or efforts targeted to low performers. One might expect highercontemplation and action scores from high-potential candidates and higherprecontemplation scores from low performers. High performers or high-potentialemployees are likely to have high-achievement motives which should incline them togreater awareness of improvement opportunities (Harris and Feild, 1992). Lowperformers are more likely to be unaware of a need for improvement or make externalattributions to explain their poor performance to avoid ego-damaging dissonance. Ifthe above assumptions are true, they suggest that different developmental strategieswould be required for high versus low performers. Unfortunately, the present study didnot allow us to differentiate participants by performance. Future research wouldbenefit from such differentiation.

The developmental effort we studied relied solely on traditional classroom training.There are many more approaches to leadership development that deserve attention(Goldstein and Ford, 2002). In fact, in the realm of leadership development, classroomtraining is increasingly regarded as least influential while actual experience (e.g.on-the-job training) is considered the most influential (McCall, 2004). An advantage ofactual experience as a developmental tool, with regard to the stages of change model,might be its utility across the stages of change. Experiential trials and difficultiesmight develop the awareness of needs for change in precontemplators. In addition, theexperience might provide the motivation to improve for those in contemplation and theplatform for practice for those in the action stage.

Our inability to collect actual development performance data are another weakness.The literature (Colquitt et al., 2000; Goldstein and Ford, 2002; Kirkpatrick, 1994)identifies four training/developmental outcomes: learning, reactions, knowledgetransfer, and job performance. In the present study, we were only able to assess themost subjective of these outcomes, reactions to the training. Clearly, learning, transfer,and job performance are key outcomes of any leadership development program. Whilewe would expect that these outcomes would be lowest for precontemplators andhighest for contemplators and those in the action stages, data are needed to make such

Stages of changeperspective

789

an assessment and future research would benefit from objective performance datacollected longitudinally.

Another potential weakness concerns the exclusive use of self-report data. Althoughsome have questioned the influence of common-method variance (Crampton andWagner, 1994; Harrison et al., 1996) and Spector (1994) has noted that cross-sectionaldesigns can be useful during the early stages of concept development, we cannot ruleout the possibility that common-method variance influenced the reported correlationsbetween participants’ SOCS-scores and their learning orientations, organizationalattitudes, and perceptions of their developmental need areas. It should be borne inmind, however, that the focus of the current study was on the pattern of relationshipsbetween different variables, and not just their magnitude. Collection of other trainingoutcomes as suggested earlier would also help with common methods problems,assuming alternative assessments to self-reports are employed as a part of the trainingand development activities.

A final weakness of the study is the lack of reported demographic information. Thishas made it impossible to determine if there are significant differences due to sex,ethnicity, education, age, etc. We recognize the importance of such information;however, given the confidentiality arrangements made with the company we could notcollect it. We recommend the inclusion of demographic information by futureresearchers to further our understanding of the stages of change model in leadershipand other organizational contexts.

Despite these limitations, the present investigation is unique in that makes use of a“stages of change” model, derived from the therapeutic literature, to empiricallyexamine differential patterns of relationships between the stages of change andparticipants’ organizational attitudes, leadership developmental needs, andlongitudinal reactions to the development effort. Consistent with Cole et al. (2004),our study provides support for the appropriateness of applying the stages of changemodel and its measurement in a management development context. We feel suchapplication has the potential for guiding pragmatic efforts to tailor developmentprograms and pre-development readiness/awareness efforts to meet the diverse needsof management participants with diverse stage sentiments and correspondingmotivations.

References

Allen, N.J. and Meyer, J.P. (1990), “The measurement and antecedents of affective, continuance,and normative commitment to the organization”, Journal of Occupational Psychology,Vol. 63, pp. 1-18.

Alliger, G.M., Tannenbaum, S.I., Bennett, W., Traver, H. and Shotland, A. (1997),“A meta-analysis of the relations among training criteria”, Personnel Psychology, Vol. 50,pp. 341-58.

Armenakis, A.A., Harris, S.G. and Mossholder, K.W. (1993), “Creating readiness fororganizational change”, Human Relations, Vol. 46, pp. 681-703.

Armenakis, A.A., Harris, S.G. and Feild, H.S. (1999), “Making change permanent:institutionalizing change interventions”, in Pasmore, W.A. and Woodman, R.W. (Eds),Research in Organizational Development and Change,Vol. 12, JAI Press, Stamford, CT,pp. 97-128.

JOCM20,6

790

Baldwin, T.T. and Ford, J.K. (1988), “Transfer of training: a review and directions for futureresearch”, Personnel Psychology, Vol. 41, pp. 63-105.

Baldwin, T.T., Magjuka, R.J. and Loher, B.T. (1991), “The perils of participation: effects of choiceof training on trainee motivation and learning”, Personnel Psychology, Vol. 44, pp. 51-65.

Blau, P. (1964), Exchange and Power in Social Life, Wiley, New York, NY.

Cole, M.S., Harris, S.G. and Feild, H.S. (2004), “Stages of learning motivation: development andvalidation of a measure”, Journal of Applied Social Psychology, Vol. 34, pp. 1421-56.

Colquitt, J.A. and Simmering, M.J. (1998), “Conscientiousness, goal orientation, and motivation tolearn during the learning process: a longitudinal study”, Journal of Applied Psychology,Vol. 83, pp. 654-65.

Colquitt, J.A., LePine, J.A. and Noe, R.A. (2000), “Toward an integrative theory of trainingmotivation: a meta-analytic path analysis of 20 years of research”, Journal of AppliedPsychology, Vol. 85, pp. 678-707.

Crampton, S.M. and Wagner, J.A. III (1994), “Percept-percept inflation in microorganizationalresearch: an investigation of prevalence and effect”, Journal of Applied Psychology, Vol. 79,pp. 67-76.

Cronbach, L.J. and Snow, R.E. (1977),Aptitudes and InstructionalMethods, Irvington, New York, NY.

Cunningham, C.E., Woodward, C.A., Shannon, H.S., MacIntosh, J., Lendrum, B., Rosenbloom, D.and Brown, J. (2002), “Readiness for organizational change: a longitudinal study ofworkplace, psychological and behavioural correlates”, Journal of Occupational &Organizational Psychology, Vol. 75, pp. 377-92.

DiClemente, C.C. and Prochaska, J.O. (1982), “Self-change and therapy change of smokingbehavior: a comparison of processes of change in cessation and maintenance”, AddictiveBehavior, Vol. 7, pp. 133-42.

Eisenberger, R., Huntington, R., Hutchinson, S. and Sowa, D. (1986), “Perceived organizationalsupport”, Journal of Applied Psychology, Vol. 71, pp. 500-7.

Goldstein, I.L. and Ford, J.K. (2002), Training in Organizations, 4th ed., Thomson Learning,Wadsworth.

Gouldner, A.W. (1960), “The norm of reciprocity: a preliminary statement”, AmericanSociological Review, Vol. 25, pp. 53-62.

Harris, S.G. and Feild, H.S. (1992), “Realizing the ‘potential’ of ‘high-potential’ managementdevelopment programs”, The Journal of Management Development, Vol. 11, pp. 61-70.

Harrison, D.A., McLaughlin, M.E. and Coalter, T.M. (1996), “Context, cognition, and commonmethod variance: psychometric and verbal protocol evidence”, Organizational Behaviorand Human Decision Processes, Vol. 68, pp. 246-61.

Kirkpatrick, D.L. (1994), Evaluating Training Programs: The Four Levels, Berrett-Koehler,San Francisco, CA.

Kraiger, K., Ford, J.K. and Salas, E. (1993), “Application of cognitive, skill-based, and affectivetheories of learning outcomes to new methods of training evaluation”, Journal of AppliedPsychology, Vol. 78, pp. 311-28.

Lam, C.S., McMahon, B.T., Priddy, D.A. and Gehred-Schultz, M.A. (1988), “Deficit awareness andtreatment performance among traumatic head injury adults”, Brain Injury, Vol. 2,pp. 235-42.

Lam, C.S., Chan, F. and McMahon, B.T. (1991), “Factorial structure of the change assessmentquestionnaire for individuals with traumatic head injuries”, Rehabilitation Psychology,Vol. 36, pp. 189-99.

Stages of changeperspective

791

Levesque, D.A., Prochaska, J.M., Prochaska, J.O., Dewart, S.R., Hamby, L.S. and Weeks, W.B.(2001), “Organizational stages and processes of change for continuous qualityimprovement in health care”, Consulting Psychology Journal: Practice and Research,Vol. 53, pp. 139-53.

McCall, M.W. (2004), “Leadership development through experience”, Academy of ManagementExecutive, Vol. 18, pp. 127-30.

McConnaughy, E.A., Prochaska, J.O. and Velicer, W.F. (1983), “Stages of change inpsychotherapy: measurement and sample profiles”, Psychotherapy: Theory, Researchand Practice, Vol. 20, pp. 368-75.

McConnaughy, E.A., DiClemente, C.C., Prochaska, J.O. and Velicer, W.F. (1989), “Stages ofchange in psychotherapy: a follow-up report”, Psychotherapy, Vol. 26, pp. 494-503.

Mathieu, J.E., Tannenbaum, S.I. and Salas, E. (1992), “Influences of individual and situationalcharacteristics on measures of training effectiveness”, Academy of Management Journal,Vol. 35, pp. 828-47.

Meyer, J.P. and Allen, N.J. (1984), “Testing the ‘side-bet theory’ of organizational commitment:some methodological considerations”, Journal of Applied Psychology, Vol. 69, pp. 372-8.

Morera, O.F., Johnson, T.P., Freels, S., Parsons, J., Crittenden, K.S., Flay, B.R. and Warnecke, R.B.(1998), “The measure of stage of readiness to change: some psychometric considerations”,Psychological Assessment, Vol. 10, pp. 182-6.

Noe, R.A., Wilk, S.L., Mullen, E.J. and Wanek, J.E. (1997), “Employee development: constructvalidation issues”, in Ford, J.K. (Ed.), Improving Training Effectiveness in WorkOrganizations, Erlbaum, Mahwah, NJ, pp. 153-89.

Paulhus, D.L. (1991), “Measurement and control of response bias”, in Robinson, J.P., Shaver, P.R.and Wrightsman, L.S. (Eds), Measures of Personality and Social Psychological Attitudes,Academic Press, San Diego, CA, pp. 17-59.

Prochaska, J.M., Levesque, D.A., Prochaska, J.O., Dewart, S.R. and Wing, G.R. (2001a),“Mastering change: a core competency for employees”, Brief Treatment and CrisisIntervention, Vol. 1, pp. 7-15.

Prochaska, J.M., Prochaska, J.O. and Levesque, D.A. (2001b), “A transtheoretical approach tochanging organizations”, Administration and Policy in Mental Health, Vol. 28, pp. 247-61.

Prochaska, J.M., Paiva, A.L., Padula, J.A., Prochaska, J.O., Montgomery, J.E., Hageman, L. andBergart, A.M. (2005), “Assessing emotional readiness for adoption using thetranstheoretical model”, Children and Youth Services Review, Vol. 27, pp. 135-52.

Prochaska, J.O. (1979), Systems of Psychotherapy: A Transtheoretical Analysis, Dorsey,Homewood, IL.

Prochaska, J.O. and DiClemente, C.C. (1982), “Transtheoretical therapy: toward a moreintegrative model of change”, Psychotherapy: Theory, Research and Practice, Vol. 19,pp. 276-89.

Prochaska, J.O. and DiClemente, C.C. (1983), “Transtheoretical therapy: toward a moreintegrative model of change”, Psychotherapy: Theory, Research and Practice, Vol. 20,pp. 161-73.

Prochaska, J.O., DiClemente, C.C. and Norcross, J.C. (1992), “In search of how people change:applications to addictive behaviors”, American Psychologist, Vol. 47, pp. 1102-14.

Prochaska, J.O., Velicer, W.F., Rossi, J.S., Goldstein, M.G., Marcus, B.H., Rakowski, W., Fiore, C.,Harlow, L.L., Redding, C.A., Rosenbloom, D. and Rossi, S.R. (1994), “Stages of change anddecisional balance for 12 problem behaviors”, Health Psychology, Vol. 13, pp. 39-46.

JOCM20,6

792

Spector, P.E. (1994), “Using self-report questionnaires in OB research: a comment on the use of acontroversial method”, Journal of Organizational Behavior, Vol. 15, pp. 385-92.

Spector, P.E. and Brannick, M.T. (1995), “The nature and effects of method variance inorganizational research”, International Review of Industrial and Organizational Psychology,Vol. 10, pp. 249-74.

Tannenbaum, S.I. and Yukl, G. (1992), “Training and development in work organizations”,Annual Review of Psychology, Vol. 43, pp. 399-441.

Tellegen, A. (1982), “Brief manual for the differential personality questionnaire”, unpublishedmanuscript, University of Minnesota, Minneapolis, MN.

Tellegen, A. (1985), “Structure of mood and personality and their relevance to assessing anxiety,with an emphasis on self-report”, in Tuma, A.H. and Maser, J.D. (Eds), Anxiety and theAnxiety Disorders, Erlbaum, Hillsdale, NJ, pp. 681-706.

Thompson, C., Koon, E., Woodwell, W.H. and Beauvais, J. (2002), “Training for the next economy:an ASTD state of the industry report on the trends in employer-provided training in theUnited States”, Report conducted by the American Society of Training & Development.

Tichy, N.M. and Cohen, E. (1997), The Leadership Engine: How Winning Companies BuildLeaders at Every Level, Harper Collins Publishers, New York, NY.

Tichy, N.M. and Sherman, S. (1993), Control Your Own Destiny or Someone Else Will, Doubleday,New York, NY.

VandeWalle, D.M. (1996), “Are our students trying to prove or improve their ability?Development and validation of an instrument to measure academic goal orientation”,paper presented at the 1996 Academy of Management National Meeting, Cincinnati, OH,August.

Velicer, W.F., DiClemente, C.C., Prochaska, J.O. and Brandenburg, N. (1985), “Decisional balancemeasure for assessing and predicting smoking status”, Journal of Personality and SocialPsychology, Vol. 48, pp. 1279-89.

Corresponding authorStanley G. Harris can be contacted at: harris@business.auburn.edu

Stages of changeperspective

793

To purchase reprints of this article please e-mail: reprints@emeraldinsight.comOr visit our web site for further details: www.emeraldinsight.com/reprints